Mehthod for image conversion

Abstract

A method is proposed for image conversion of image data with a first contrast range to image data with a second contrast range. Fourier coefficients of a Fourier transform of the image data are entered in a neural network by calculating parameters for carrying out windowing.

1 . A method for image conversion of image data with a first contrast range to image data with a second contrast range via windowing, the method comprising: 
 determining at least one input parameter for a neural network from the image data, with a Fourier transformation being carried out during the determination process;    entering the at least one input parameter in the neural network; and    calculating a center and a width of a window, for use by the neural network.    
   
   
       2 . The method as claimed in  claim 1 , further comprising: 
 removing, before the image data is entered in the neural network, an image background which does not contribute to relevant image information.    
   
   
       3 . The method as claimed in  claim 1 , wherein the image data is scaled to a previously defined image size before being entered in the neural network.  
   
   
       4 . The method as claimed in  claim 1 , wherein the Fourier transformation is used to calculate a Fourier transform of the image data.  
   
   
       5 . The method as claimed in  claim 1 , wherein a histogram of the image data is calculated before the Fourier transformation, from which a Fourier transform is then calculated.  
   
   
       6 . The method as claimed in  claim 5 , wherein a selection of Fourier coefficients of the Fourier transforms is entered in the neural network, for processing.  
   
   
       7 . The method as claimed in  claim 6 , wherein numerical values are entered in the neural network for processing, which numerical values are functionally related to the Fourier coefficients of the Fourier transforms.  
   
   
       8 . The method as claimed in  claim 1 , wherein the data which has been entered in the neural network is processed by an input level, a concealed level and an output level.  
   
   
       9 . The method as claimed in  claim 8 , wherein thirty six input neurons in the neural network transmit the entered data via weighted connections to twelve neurons in the concealed level, and the twelve neurons in the concealed level transmit the data via weighted connections to two output neurons in the output level.  
   
   
       10 . The method as claimed in  claim 8 , wherein nineteen input neurons in the input level in the neural network transmit the entered data via weighted connections to twelve neurons in the concealed level, and the twelve neurons in the concealed level transmit the data via weighted connections to two output neurons in the output level.  
   
   
       11 . The method as claimed in  claim 2 , wherein the image data is scaled to a previously defined image size before being entered in the neural network.  
   
   
       12 . The method as claimed in  claim 4 , wherein a histogram of the image data is calculated before the Fourier transformation, from which a Fourier transform is then calculated.  
   
   
       13 . The method as claimed in  claim 4 , wherein a selection of Fourier coefficients of the Fourier transforms is entered in the neural network, for processing.  
   
   
       14 . The method as claimed in  claim 9 , wherein nineteen input neurons in the input level in the neural network transmit the entered data via weighted connections to twelve neurons in the concealed level, and the twelve neurons in the concealed level transmit the data via weighted connections to two output neurons in the output level.  
   
   
       15 . A computer program, adapted to, when executed on a computer device, cause the computer device to carry out the method as claimed in  claim 1 .  
   
   
       16 . A computer readable medium, including the computer program of  claim 15 .  
   
   
       17 . A method for image conversion of image data with a first contrast range to image data with a second contrast range, the method comprising: 
 determining a window, at least one input parameter for a neural network being initially determined, utilizing Fourier transformation, from the image data, and the window being determined from the neural network including the at least one input parameter; and    converting image data with the first contrast range into image data with the second contrast range using the determined window.    
   
   
       18 . The method as claimed in  claim 1 , wherein the image data is scaled to a previously defined image size before being entered in the neural network.  
   
   
       19 . A computer program, adapted to, when executed on a computer device, cause the computer device to carry out the method as claimed in  claim 17 .  
   
   
       20 . A computer readable medium, including the computer program of  claim 19.